Of the estimated 168,811 men who will be treated for prostate cancer (PC) with curative intent in 2006, >30% will ultimately experience disease recurrence, metastasis, and die of PC. There is substantial uncertainty, however, in predicting which individual patients will have adverse outcomes based on both clinical presentation and pathological features. In the previous proposal, we hypothesized the complexity of the PC progression process could be explained by genetic variants that confer susceptibility to more aggressive or to lethal PC. That is, subtle changes in transcription,translation or the function of gene products due to inherited genetic variants may modify tumor behavior. In the current proposal, we will expand our investigation into whether allelic variation plays a role in metastatic efficiency (i.e.. the ability of a tumor to spread or recur), with the following specific aims: 1) Evaluate the association of genetic polymorphisms in candidate genes with PC-specific mortality and disease recurrence/progression using two population-based cohorts of PC patients (follow-up mean=10.4 yrs.;range 5-17 yrs.). As part of the initial SPORE proposal a panel of 384 single nucleotide polymorphisms (SNPs) in candidate genes associated with PC pathways (e.g., androgens) has been genotyped in Cohort 1 (n=630). For the current proposal, interesting SNPs (i.e., p<0.15 in relation to PC-specific death from Cox models) from Cohort 1 will be genotyped in patient Cohort 2 (n=840) for confirmation. SNPs that are validated will be examined with outcomes in both cohorts; 2) Examine tagSNPs in PC metastasis-suppressor genes (e.g., KAI1, MAPKK4) in relation to PCoutcomes as defined in Aim 1 in all patients with DNA from both cohorts (n=1,470);and, 3) Evaluate genotype-phenotype associations within the combined cohort of PC patients to identify SNPs that are correlated with more aggressive or less aggressive disease features. These analyses will consider whether individual SNPs, haplotypes (tagSNPs within candidate genes), or groups of SNPs in candidate genes in defined pathways are independently associated with PC aggressiveness. We anticipate this study will provide new insights into genetic pathways and mechanisms that determine metastatic potential, pushing PC toward its lethal phenotype. Knowledge of metastasis-modifier alleles may help identify subsets of PC patients at higher risk for adverse outcomes, those who would benefit most from tailored individual therapies (e.g., early adjuvant therapy, or expectant management), from heightened surveillance for recurrence/progression, and early interventions aimed at reducing PC deaths.